PostgreSQL Query Catalog

What is the percentage of optional columns (that permit NULLs) in case of each base table? It is better to prohibit the use of NULLs in as many columns as possible. Otherwise the results of queries may be misleading.

This query identifies table columns and domains that are configured with a semantically void DEFAULT value. It specifically flags defaults that are an empty string ('') or a string consisting solely of whitespace characters (e.g., spaces, newlines). This practice is a design flaw because it automatically populates the database with non-substantive data, which can lead to application-level bugs when code does not explicitly check for such "blank" values in addition to NULL.

Find base tables where more than one column gets the default value by using the sequence generator mechanism.

Find implementations of state machines that uses a set of columns with a timestamp type.

Find base table columns with Boolean type that name refers to the possibility that these are used to register as to whether an entity is currently in active state or not. Find the base tables that have exactly one Boolean column. During the system design one should find all the possible states of an entity type that influence the behavior of the information system. Data as to whether an entity is in one of these states should be in the database. Having only two states - active/inactive - is sometimes a too big simplification.

Find check constraints of tables that use a regular expression to check as to whether a registered string ends with an appropriate file extension. However, the expression does not put the dot sign into the square brackets nor does have the escape character \before it, i.e., it is interpreted as a single character not as the dot sign in the expression. In regular expressions the dot (.) matches any single character except the newline character.

This query identifies a semantic mismatch between column definitions and their identifiers. It flags base table columns that are configured with a DEFAULT value of CURRENT_USER or SESSION_USER (indicating they store user identity) but whose names fail to reflect this purpose. Specifically, it searches for columns lacking semantic cues such as "user", "login", "owner", or "by" in their names. This obscuration reduces schema self-documentation, as developers cannot intuitively determine that the column is intended for audit or ownership tracking.

Find columns of base tables that have default value CURRENT_USER.

Find row level triggers that have action condition (WHEN clause) but the Boolean expression in its specifications does not refer to neither NEW nor OLD variable.

Use a right predicate in trigger condition in order to ensure that the trigger executes always when it has to but not more often. IS DISTINCT FROM treats NULL as if it was a known value, rather than unknown. It would be relevant if a column that is referenced in the action condition is optional, i.e., permits NULLs.

Find base table columns that do not have uuid type but the name of the column refers to the possibility that the values in the column are uuid's.

Make sure that in case of using the type DECIMAL/NUMERIC as the type of a base table column the precision (the permitted number of digits in the number) is not too small. For instance, the biggest value in the type NUMERIC(1,1) is 0.9.

Find base table columns and domains that have a textual type and the default value that represents a truth-value. For instance, the type of a column could be VARCHAR and the column has the default value 'TRUE'.

Specify for each column/domain a right data type that takes into account expected values in the column/domain. Find base table columns and domains that refer to the nextval function by using the default value mechanism but do not have the type INTEGER, SMALLINT, or BIGINT. This check is performed in case of identity columns: ERROR: identity column type must be smallint, integer, or bigint.

Specify for each column/domain a right data type that takes into account expected values in the column/domain. Find base table columns and domains that have a textual type but the default value that represents a number (for instance, '100', '2', or '0.22'). Exclude columns about formats.

Find columns that name points to the possibility that values in this are classifier codes. The column has a numeric type but it is not SMALLINT. Usually each classifier type has so few values that type SMALLINT would be appropriate.

Find base table columns and domains that have a textual type but the default value that represents a temporal value (either a static value or invocation of a function that returns such value). Specify for each column/domain a right data type that takes into account expected values in the column/domain.

Find base table columns that name refers to the possibility that there are registered deadlines but the column does not have a temporal type.

Find base table columns that have CHAR type, where character maximum length is bigger than 1 and the name of the column does not refer to the possibility that the column holds some kind of codes or flags or hash values.

Find columns of base tables where the name of the column has prefix or suffix "id" or has the name "id" but the column does not have an integer type or uuid type. A convention is to use the phrase "id" in the names of surrogate key columns.